libera.catirclogs.org

michaelni changed the topic of #ffmpeg-devel to: Welcome to the FFmpeg development channel | Questions about using FFmpeg or developing with libav* libs should be asked in #ffmpeg | This channel is publicly logged | FFmpeg 9.0.1 has been released! | Please read ffmpeg.org/developer.html#Code-of-conduct

<fjlogger> [FFmpeg/FFmpeg] Pull request #24273 opened: [release/4.4] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24273) by f⁠fmpeg-devel

<fjlogger> [FFmpeg/FFmpeg] Pull request #24274 opened: [release/5.1] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24274) by f⁠fmpeg-devel

<fjlogger> [FFmpeg/FFmpeg] Pull request #24275 opened: [release/6.1] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24275) by f⁠fmpeg-devel

<fjlogger> [FFmpeg/FFmpeg] Pull request #24276 opened: [release/7.1] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24276) by f⁠fmpeg-devel

<fjlogger> [FFmpeg/FFmpeg] Pull request #24277 opened: [release/8.0] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24277) by f⁠fmpeg-devel

<fjlogger> [FFmpeg/FFmpeg] Pull request #24278 opened: [release/8.1] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24278) by f⁠fmpeg-devel

<fjlogger> [FFmpeg/FFmpeg] Pull request #24279 opened: [release/9.0] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24279) by f⁠fmpeg-devel

<kasper93> I honestly don't know why ffmpeg has 7 supported release branches

<kasper93> it should be one or two LTS plus current release

<kasper93> BtbN: I think there is some opportunities to make dispatch to GHA better. I see that GHA spawned 9 ubuntu (at one action), but most of them are idle

<BtbN> If you have any idea how

<BtbN> It just checks how many jobs are waiting, and fires up that many runners

<BtbN> After 30 seconds

<BtbN> The idea being "rather too many than too few"

<BtbN> Where do you even see idle runners? They all seem pretty dang busy to me

<kasper93> but it also spawned 5 actions, with varying number of jobs, instead of connecting to existing idle runners. It looks like race to start action

<BtbN> There never are idle runners on GHA

<BtbN> They pull for one job, and then exit, and end the GHA workflow

<BtbN> no waiting

<BtbN> There's a slight optimization in there in that after a runner on there has successfully processes a job, it will poll again one more time

<BtbN> to skip the whole spinup and work through jobs faster

<kasper93> I see

<BtbN> Which can lead to a bit of a "tail" of workers, that then will not find a job and just exit

<BtbN> But the GHA "bridge" can't ever have idle runners

<kasper93> I though maybe the "run runner" waits, because the numbers didn't add up

<kasper93> most jobs were waiting of forgejo, while GHA had 20+ spawned jobs

<kasper93> but it looks like the spin up alone takes 3+ minutes

<BtbN> We have simply maxed out GHA right now

R-Camacho has quit [Ping timeout: 241 seconds]

<BtbN> it has 20 runners working on Forgejo jobs

<BtbN> Yes, the spinup is quite slow

<kasper93> so, it was just waiting for the workers to be ready to actuall report progress to forgejo

<BtbN> But right now every single worker I click on is working through jobs

minimal has quit [Quit: Leaving]

<kasper93> all good then

<BtbN> most of the spinup time is restoring docker images from the cache

<BtbN> Not great, but still faster than downloading all of them fresh every single run

<fjlogger> [FFmpeg/FFmpeg] Pull request #24273 closed: [release/4.4] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24273) by k⁠asper93

Compnn has quit [Read error: Connection reset by peer]

Compn has joined #ffmpeg-devel

___nick___

has joined #ffmpeg-devel

___nick___

has quit [Client Quit]

___nick___

has joined #ffmpeg-devel

<fjlogger> [FFmpeg/FFmpeg] Pull request #24274 merged: [release/5.1] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24274) by k⁠asper93

<fjlogger> [FFmpeg/FFmpeg] Pull request #24275 merged: [release/6.1] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24275) by k⁠asper93

<fjlogger> [FFmpeg/FFmpeg] Pull request #24276 merged: [release/7.1] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24276) by k⁠asper93

<fjlogger> [FFmpeg/FFmpeg] Pull request #24277 merged: [release/8.0] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24277) by k⁠asper93

<fjlogger> [FFmpeg/FFmpeg] Pull request #24278 merged: [release/8.1] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24278) by k⁠asper93

<fjlogger> [FFmpeg/FFmpeg] Pull request #24279 merged: [release/9.0] avcodec/h264_direct: do not read the other field's rows of colocated mb_type (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/24279) by k⁠asper93

boplaid has quit [Quit: byebye]

averne_ has joined #ffmpeg-devel

averne has quit [Ping timeout: 268 seconds]

averne_ is now known as averne

thardin has quit [Server closed connection]

thardin has joined #ffmpeg-devel

witchymary has quit [Ping timeout: 260 seconds]

witchymary has joined #ffmpeg-devel

akinji_ has quit [Ping timeout: 243 seconds]

witchymary has quit [Read error: Connection reset by peer]

witchymary has joined #ffmpeg-devel

av500 has joined #ffmpeg-devel

_av500_

has quit [Ping timeout: 270 seconds]

jamrial has quit []

odrling has quit [Server closed connection]

odrling has joined #ffmpeg-devel

agentcasey has quit [Server closed connection]

jarthur_ has joined #ffmpeg-devel

jarthur_ has quit [Client Quit]

jarthur has quit [Ping timeout: 253 seconds]

jarthur has joined #ffmpeg-devel

Martchus has joined #ffmpeg-devel

Martchus_ has quit [Ping timeout: 276 seconds]

pross has joined #ffmpeg-devel

kylophone has quit [Server closed connection]

kylophone has joined #ffmpeg-devel

Fenrir has quit [Server closed connection]

Fenrir has joined #ffmpeg-devel

Nightrose has quit [Server closed connection]

Nightrose has joined #ffmpeg-devel

Compn has quit [Remote host closed the connection]

Compn has joined #ffmpeg-devel

mateo` has quit [Server closed connection]

mateo` has joined #ffmpeg-devel

jos has joined #ffmpeg-devel

APic has quit [Server closed connection]

APic has joined #ffmpeg-devel

Read the original on libera.catirclogs.org ↗